Sandra Daniela Silva
Guedes
Avaliação de polimorfismos em genes associados a
inflamação num grupo de risco para demência
Testing a dementia risk group for polymorphisms in
inflammation-related genes
2016
Sandra Daniela Silva
Guedes
Avaliação de polimorfismos em genes associados
a inflamação num grupo de risco para demência
Testing a dementia risk group for polymorphisms
in inflammation-related genes
Tese apresentada à Universidade de Aveiro para cumprimento dos requisitos necessários à obtenção do grau de Mestre em Biomedicina Molecular, realizada sob a orientação científica da Professora Doutora Ana Gabriela Henriques, Professora Auxiliar Convidada do Departamento de Ciências Médicas da Universidade de Aveiro.
Este trabalho foi desenvolvido no grupo de Neurociências e Sinalização, e contou com o apoio do Instituto de Biomedicina Molecular (iBiMED), UID/BIM/04501/2013, da Universidade de Aveiro e da FCT (projeto PTDC/DTP-PIC/5587/2014 e JPND-BIOMARKAPD).
o júri
presidente Doutora Sandra Maria Tavares da Costa Rebelo
Professora Auxiliar Convidada da Universidade de Aveiro
Doutora Sónia Cristina das Neves Ferreira
Investigadora de Pós-Doutoramento do Centro de Neurociências e Biologia Celular da Universidade de Coimbra
Doutora Ana Gabriela da Silva Cavaleiro Henriques
agradecimentos À minha orientadora, Professora Doutora Ana Gabriela Henriques, agradeço toda a disponibilidade, apoio, dedicação e ensinamentos transmitidos.O meu mais sincero e profundo agradecimento por todo o tempo perdido e pela paciência prestada durante o decorrer deste trabalho.
À Professora Doutora Odete da Cruz e Silva pela oportunidade de participar neste projeto, assim como pelo encorajamento e motivação transmitida. À Professora Doutora Ilka Martins Rosa pela colaboração neste projeto. Agradeço aos voluntários deste estudo e famílias, bem como todos os profissionais envolvidos que possibilitaram a realização deste estudo. A todos os meus colegas do Laboratório de Neurociências, por todos os conselhos, pela amizade, pelo apoio e ajuda sempre disponibilizados e por todos os sábios ensinamentos que partilharam comigo ao longo desta minha etapa.
A todos os meus colegas de mestrado, em especial à Marlene e à Tânia pelo companheirismo e apoio demostrado ao longo desta minha estadia em Aveiro.
À Maria, com quem tive o prazer de partilhar esta nova etapa em Aveiro. Por ter sido o meu apoio, pela preocupação mostrada, por todos os bons momentos vividos e acima de tudo por me ter mostrado o verdadeiro significado de amizade e por ter estado sempre presente, mesmo apesar da distância.
Aos meus pais e ao meu irmão, por toda a força, carinho, coragem e confiança que me transmitiram. Obrigada por todo o apoio e por estarem sempre presentes em todos os momentos da minha vida, por me ensinarem a contornar os obstáculos, e acima se tudo, a ser feliz. Porque tudo o que tenho é a vocês que o devo. Muito obrigada.
palavras-chave Doença de Alzheimer; neuroinflamação; Polimorfismos genéticos; Clusterina; Receptor do complemento 1.
resumo A doença de Alzheimer (AD) é uma perturbação degenerativa multifatorial associada com a idade que ocorre no sistema nervoso central. Após a sua descrição inicial em 1907, numerosas teorias foram propostas para elucidar quais as principais causas associadas. A hipótese da inflamação tem sido recentemente reconhecida pela comunidade científica, uma vez que muitos estudos em modelos e doentes de Alzheimer propuseram fortes evidências da ativação do sistema imunológico e de processos inflamatórios durante o curso da doença. De facto, a acumulação de β-amilóide (Aβ) e proteína tau provocam uma resposta inflamatória cerebral como resultado do desenvolvimento patológico da AD. Atualmente, os estudos de associação genómica genética (GWAS) proporcionaram a identificação de diversas variantes genéticas que influenciam por exemplo processos inflamatórios e as vias do sistema imunitário na AD, estando as regiões polimórficas CLU rs11136000 e CR1 rs3818361 entre elas. Além disso, ambos os polimorfismos de um único nucleótido (SNPs) parecem ter um papel colaborativo relativamente à eliminação de Aβ e à ativação do sistema imunitário através da estimulação do complemento. No trabalho aqui descrito, foram realizadas análises bioinformáticas de genes de risco para a AD, principalmente o CLU e o CR1. As informações obtidas foram usadas para criar uma rede de interação proteína-proteína, bem como para realizar análises de enriquecimento de Ontologia Genética. A nossa análise bioinformática indica que ambos os genes CLU e CR1 estão envolvidos numa variedade de vias de sinalização que compreendem a regulação do processo inflamatório e ativação do sistema imunológico.
A expressão genética de cada alelo de risco das SNPs CLU rs11136000 e CR1 rs3818361 foi ainda avaliada em amostras de doentes “Putativos AD” e Controlos por testes de PCR e análises de sequenciação de Sanger. Adicionalmente, as frequências genotípicas e alélicas também foram determinadas com o intuito de criar um perfil genético dos grupos estudados. Os nossos resultados demostraram que no grupo de doentes “Putativos AD” analisado para a variante CLU rs11136000, o alelo de risco C apresentou maior frequência (64%) quando comparado com o grupo Controlos (40%). O grupo de Controlos apresentou uma frequência de 60% para o alelo de não-risco. Para a variante CR1 rs3818361, o alelo de risco A apresentou frequências semelhantes entre grupos, apesar do aumento da percentagem de homozigóticos de risco (6%) no grupo de doentes “Putativos AD”.
Este trabalho auxilia na compreensão da relação entre estes polimorfismos genéticos e demência. Estudos adicionais devem avaliar o uso destas SNPs como ferramentas potencialmente úteis no diagnóstico da AD.
keywords Alzheimer’s disease; neuroinflamamtion; Genetic polymorphisms; Clusterin; Complement-receptor 1.
abstract Alzheimer’s disease (AD) is a multifactorial age associated degenerative disorder that occurs in the central nervous system. After its initial report in 1907, numerous theories have been proposed to elucidate on what are the related main causes. The inflammation hypothesis has been recently acknowledged by the scientific community since several studies in AD models and patients strongly supported the activation of the immune system and of inflammatory processes during disease development. In fact, the accumulation of amyloid-β (Aβ) and tau-neurofibrillary tangles provokes a brain inflammatory response as a consequence of the pathological development of AD. Currently, genome-wide association studies (GWAS) have provided several genetic variants that impact inflammation and immune system pathways in AD, being the polymorphic regions CLU rs11136000 and CR1 rs3818361 among them. Furthermore, both single-nucleotide polymorphisms (SNPs) appear to have a collaborative role regarding Aβ clearance and immune system activation via complement stimulation.
In the work here described, bioinformatics analyses of AD risk-related genes, focusing on CLU and CR1 were performed and the retrieved information used to rise a protein-protein interaction network, as well as to perform Gene Ontology enrichment analyses. Our bioinformatics analysis indicates that CLU and CR1 are involved in a variety of signaling pathways that comprise activation and regulation of immune system process.
CLU rs11136000 and CR1 rs3818361 genetic expression of each SNP risk allele was further evaluated in whole blood samples from “Putative AD” and Controls groups by PCR assays and Sanger sequencing analyses. Additionally, the genotyping and allelic frequencies were also determined in order to create a genetic profile of the studied groups.
Our results showed that on the “Putative AD” group analyzed for CLU rs11136000 variant, the C-risk allele presented a higher frequency (64%) when compared to Controls (40%). The Controls group displayed a 60% frequency for the non-risk allele. For the CR1 rs3818361 variant, the A-risk allele showed similar frequencies among groups, although an increase in the percentage of homozygous risk carriers (6%) was observed in the “Putative AD” group. This work aids into the understanding of the relation between these genetic polymorphisms and dementia. Additional studies should address the use of these SNPs as potential tools in AD diagnostics.
LIST OF CONTENTS
Abbreviations
... iii
Introduction
... 1
1.1 Alzheimer’s disease and its neuropathological hallmarks ... 1
1.2 Genetics of Alzheimer’s disease and Genome-wide association studies (GWAS)
..4
1.3 Inflammation and immune response in Alzheimer’s disease ... 8
1.3.1 Aβ and Neuroinflammation………..8
1.3.2 Immunity effects………13
1.4 Clusterin ... 15
1.4.1 Clusterin in Alzheimer’s disease……….………..…….15
1.5 Complement receptor-1 ………19
1.5.1 Complement receptor 1 in Alzheimer’s disease……….…………....20
Aims of the Thesis
... 23
Materials and Methods
... 25
3.1 Bioinformatics analyses ... 25
3.1.1 Literature survey on AD risk-related genes and construction of a
protein-protein interaction network…..………25
3.1.2 Gene ontology analysis………...……..….25
3.2 Characterization of the study group……….25
3.3 PCR analyses……….……….27
3.3.1 Blood samples collection………..…..……...27
3.3.2 Genotyping test……….………27
3.3.3 Precipitation of DNA fragments………..………..28
3.3.4 Agarose gel electrophoresis………..……..…28
Results
……….……….31
4.1 Bioinformatics analyses of CLU and CR1 genes in AD pathology……….31
4.1.1 Interaction network of AD risk-related genes……….31
4.1.2 GO analysis of CLU and CR1 within the AD risk-related genes network……..34
4.2 CLU polymorphic region rs11136000 and CR1 polymorphic region rs3818361
amplification by PCR ... 37
4.3 Associating the CLU and CR1 SNP genotype with AD ... 399
4.4 Determination of CLU and CR1 SNPs genotypic frequencies ... 41
4.4.1 CLU rs11136000 SNP genotypic frequencies………..……..41
4.4.2 CR1 rs3818361 SNP genotypic frequencies……….42
4.5 Evaluation of CLU rs11136000 SNP and CR1 rs3818361 SNP allelic frequencies . 44
4.5.1 CLU rs11136000 SNP allelic frequencies……….………..44
4.5.2 CR1 rs3818361 SNP allelic frequencies……….…….46
Discussion
... 49
5.1 AD risk-related genes network ... 49
5.2 GO analysis of CLU and CR1 within the AD risk-related genes network ... 50
5.3 CLU rs11136000 profile ... 51
5.4 CR1 rs3818361 profile ... 53
Conclusions
... 55
7. Future Perspectives
……….……….…………55
References
... 57
Appendix 1
... 67
Appendix 2
... 71
Abbreviations
Aβ
Amyloid beta-peptide
AD
Alzheimer´s disease
AICD
APP intracellular domain
APP
Amyloid precursor protein
APOE
Apolipoprotein E
APOEε4
APOE allele 4ApoJ
Apolipoprotein J
CD
Cluster of differentiation
CD2AP
CD2-associated protein
CLU
Clusterin
CNS
Central nervous system
CR1
Complement receptor-1
CCR
C-C chemokine receptor type
CSF
Cerebrospinal fluid
CTF
C-terminal fragment
DNA
Deoxyribonucleic acid
EDTA
Ethylenediaminetetraacetic acid
EOAD
Early-onset familial Alzheimer’s disease
GFAP
Glial fibrillary acidic protein
GWAS
Genome-wide association study
IC
Immune complexes
IDE
Insulin-degrading enzyme
IG
Immunoglobulins
IL
Interleukin
iNOS
Nitric oxide synthase
LOAD
Late-onset Alzheimer’s disease
LB
Loading buffer
LPS
Lipopolysaccharide
MCI
Mild cognitive impairment
MRI
Magnetic resonance imaging
NaOH
Sodium Hydroxide
nCLU
Nuclear clusterin
NFTs
Neurofibrillary Tangles (NFTs)
NF-kB
Factor-nuclear kappa-B
NO
Nitric oxide
PCR
Polymerase chain reaction
PET
Positron emission tomography
PSEN1
Presenilin-1
PSEN2
Presenilin-2
PPI
Protein-protein interaction
RAGE
Advanced-glycosylation end-products
RCA
Complement activation regulators
ROS
Reactive oxygen species
SCARA
Scavenger receptor A
sCLU
Secretory clusterin
SNPs
Single nucleotide polymorphisms
TAE
Tris-acetate-EDTA
TGF
Transforming growth factor
Th
T-helper
Introduction
Alzheimer’s disease and its neuropathological hallmarks
Alzheimer’s disease (AD) is the primary form of age-related neurodegenerative dementia, which embodies an emerging global health crisis. In 2015, an estimated 46.8 million people worldwide lived with dementia, and the majority of these were diagnosed with AD (Figure 1) 1,2.
This amount will nearly double every 20 years, reaching 74.7 million in 2030 and 131.5 million in 2050. The fastest expansion in the elderly population is occurring in China, India, and their south Asian and western Pacific neighbors. Recent data showed that over 153000 people in Portugal have dementia, 90000 of AD type. In other words, at least 1% of the Portuguese population has been diagnosed with this illness, representing a striking impact on the national public health system1,3.
AD was first described in 1907 by a Bavarian psychiatrist with expertise in neuropathology named Alois Alzheimer, regarding his observations in a 51-year-old patient,
Figure 1: Expected number of people with dementia until 2050, according to the World Alzheimer Report of 2015. This analysis shows that high income countries have more affectedindividuals with dementia compared to low and middle income countries. Nevertheless, both will have a strong increase in the next 34 years. From1.
Auguste D. Her symptoms included short-term memory loss, unusual behavior and neuropathological characteristics, that later became recognized as the hallmarks of the disease4.
The initial symptoms of this disease are frequently mistaken as part of standard aging processes, or indicators of stress. With AD development, symptoms may comprise aggression, irritability, confusion, and language problems, as well as loss of long-term memory and ultimately death5.
Neuropathological alterations are represented by loss of neurons and synapses in the cerebral cortex and certain subcortical regions, resulting in gross atrophy of the affected areas and degeneration of specific brain areas, including temporal and parietal lobes and parts of the frontal cortex and cingulate gyrus5. Magnetic resonance imaging (MRI) and positron emission
tomography (PET) techniques reveal specific atrophied brain regions, predominantly in the entorhinal cortex, the hippocampus and the temporal cortex, as patients followed the typical disease development stages: from asymptomatic, to mild cognitive impairment (MCI), and finally AD6,7.
Certainly, the clinical development of AD is diverse and there are numerous features adding to the typical neuropathological extracellular senile plaques and neurofibrillary tangles (NFTs). Senile plaques, which are also known as amyloid plaques, consist of extracellular deposits of amyloid beta (Aβ) peptide, mainly represented by Aβ40, and Aβ42.
Aβ peptides derive from the proteolytic processing of amyloid precursor protein (APP), which represents an integral membrane protein with extensive expression through the body8.
Moreover, is mostly concentrated in neuronal synapses and it is associated with neurite extension and synaptic plasticity. By contrast, NFTs are intracellular lesions entailing structures of paired helical filaments composed of hyperphosphorylated tau protein8. The latter is
expressed predominantly in neurons, and it is responsible for microtubules stabilization and axonal transport (Figure 2)8.
Amyloid markers, such as cerebrospinal fluid (CSF) Aβ42 and PET amyloid tracer uptake are the primary modifications of AD progression, nonetheless they reach a plateau level by the MCI stage. On the other hand, functional and metabolic markers are identifiable by task-dependent stimulation on functional MRI and 18F-fluorodeoxyglucose PET, being unusual
indicators for the MCI stage, and continuing to alter going into the dementia stage. The final structural alterations are followed by a temporal pattern that is associated with increased tau pathology accumulation (Figure 3)7.
Figure 2: Alzheimer disease and the associated morphological alterations in the brain. This image shows gross and microscopic differences between normal and Alzheimer brain. It is possible to observe that the specific areas regarding the language and the memory are affected in the brain cross-sections in Alzheimer, compared to normal brain, which ultimately leads to cortical atrophy. Additionally, there are microscopic accumulations of intraneuronal neurofibrillary tangles, interstitial amyloid neuritic plaques and loss of neurons with gliosis in Alzheimer compared to normal brain. From9.
Genetics of Alzheimer’s disease and Genome-wide association studies
(GWAS)
Epidemiological research discovered several risk factors for AD, mainly divided into vascular (such as, smoking, obesity, dyslipidemia, diabetes, hypertension and asymptomatic cerebral infarction), psychosocial (such as lower education, poor social engagement and physical activities) and genetic factors8. The latter has, in fact, a notorious role in AD and positive family history is a
solid risk factor. Studies on identical twins established a high heritability for AD, being about 58% to 80%8.
Therefore, AD is characterized by two forms centered on genetic features, namely, early-onset familial Alzheimer’s disease (EOAD) and late-early-onset Alzheimer’s disease (LOAD)10,11,12. The
familial EOAD form is inherited by Mendelian genetic distribution in an autosomal dominant pattern and the usual age of onset is typically before 6013. The loci of causative genetic mutations include
the APP gene (located in chromosome 21) and the Presenilin-1 (PSEN1) and Presenilin-2 (PSEN2)
Figure 3: The expected evolutionary model of cognitive and biological markers of AD. Abbreviations: AD, Alzheimer disease; MCI, mild cognitive impairment; NINCDS–ADRDA, National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders Association. From 7.
genes (located on chromosomes 14 and 1, respectively), all of which have complete penetrance10.
The EOAD genes mutations and their molecular phenotypes on AD patients, such as increase Aβ formation and aggregation are summarized in Table 114.
Table 1. EOAD genes and their pathogenic effects on Alzheimer disease.
Gene Protein Chromosome Molecular Phenotype APP Amyloid-β
protein percursor
21q21 Increased Aβ42/Aβ40 ratio
Increased Aβproduction Increased Aβaggregation
PSEN1 Presenilin -1 14q24 Increased Aβ42/Aβ40 ratio
PSEN2 Presenilin -2 1q31 Increased Aβ42/Aβ40 ratio
On the other hand, LOAD covers more than 90% of AD patients, and still lacks a well-defined way of transmission15. This form shows a genetically complex pattern of inheritance of joint
mechanism between genetic risk factors, aging, environmental factors and life exposure events, that together regulate lifetime risk for AD16. Furthermore, APOE (apolipoprotein E) gene, which
encodes the APOE protein, has been established as a major risk factor for LOAD17,18and has an
important role in mediating cell signaling, synaptic plasticity, neuroinflammation (inflammation in the central nervous system (CNS)), as well as in cholesterol and lipid transport20. APOEε4 (APOE
allele 4) amplifies the risk of AD in Caucasians and Asians populations and individuals with two APOEε4 alleles have 15 to 17 times higher risk to AD than those lacking this allele21,22. Nevertheless,
the APOE genotype describes merely 20% of AD risk20.
Although EOAD mutations led to AD, APOEε4 allele represents a genetic risk factor for LOAD by increasing the age of onset in a dose-dependent manner and it is not a direct cause for AD development. Subsequently to the initial studies of APOE as a genetic risk factor for LOAD, hundreds of genes have been published with an association to AD, making it difficult to track and analyze the available data.
Currently, the best approach to search for new AD genes has been the genome-wide association study (GWAS). Several GWAS reports have identified many genes (Table S1, Appendix 1) based on numerous genetic markers as single nucleotide polymorphisms (SNPs) to find the genetic association with disease risk and/or endophenotypes such as age-of-onset, biomarkers, imaging results and neuropathological endpoints16.
These genes are associated with numerous biological processes, including cell migration, lipid transport and endocytosis, amyloidogenesis, tauopathy, synaptic and cytoskeletal function, and are also related to immunological mechanisms and neuroinflammation23 (Figure 4).
Figure 4: Rare and common variants contribute to Alzheimer’s disease risk from GWAS. This illustration shows the association between the frequency of these rare and common variants in the population obtained from GWAS with the risk of developing Alzheimer’s disease. The variants can be evaluated in a progressive scale going from low risk to medium and high risk. Additionally, the variants are illustrated in different colors according to their respective biologic function. Abbreviations: ADAM10, disintegrin and metalloproteinase domain-containing protein 10; APP, amyloid precursor protein; PS1, presenilin-1; PS2, presenilin-2; APOE4, apolipoprotein E4; BIN1, myc box-dependent-interacting protein 1; CD2AP, CD2-associated protein; EPHA1, ephrin type-A receptor 1; MS4A6A, membrane-spanning 4-domains subfamily A member 6A; CLU, clusterin; CR1, complement receptor-1; PICALM, phosphatidylinositol-binding clathrin assembly protein; ABCA7, ATP-binding cassette sub-family A member 7; CD33, myeloid cell surface antigen CD33; HLA-DRB1, HLA class II histocompatibility antigen, DRB1-1 beta chain; HLA-DRB5, HLA class II histocompatibility antigen, DR beta 5 chain; PTK2B, protein-tyrosine kinase 2-beta; SORL1, sortilin-related receptor; SLC24A4, sodium/potassium/calcium exchanger 4; RIN3, ras and rab interactor 3; DSG2, desmoglein-2; INPP5D, phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase 1; MEF2C, myocyte-specific enhancer factor 2C; NME8, thioredoxin domain-containing protein 3; ZCWPW1, zinc finger CW-type PWWP domain protein 1; NYAP1, neuronal tyrosine-phosphorylated phosphoinositide-3-kinase adapter 1; CELF1, CUGBP Elav-like family member 1; FERMT2, fermitin family homolog 2; CASS4, cas scaffolding protein family member 4; PLD3, phospholipase D3; TREM2, triggering receptor expressed on myeloid cells 2. From24.
GWAS use genotyping shared ancestral polymorphisms that typically arise in 0.5% of the overall population16. Even though a large amount of the genetic variance of LOAD continues
inexplicable, several loci containing genetic variants influencing both immunological response and inflammatory pathways in AD patients have been identified by GWAS. Among these new genes are
Clusterin (CLU) located on chromosome 8, which is a powerful regulator of complement activation
and Complement receptor-1 (CR1) placed on chromosome 1, which is a vital player on immune system. These genes both participate in Aβ clearance, and both are major players within inflammatory response that is activated as part of the brain’s innate immune system in AD (Figure 5)16,25. CLU rs11136000 and CR1 rs3818361 SNPs were reported to contribute for AD pathogenesis,
with both variants having a strong risk for LOAD26,27. CLU and CR1 involvement in AD will be further
discussed.
Figure 5: Genetic risk factors and inflammation in AD. This representation shows the association among innate immunity-related genetic risk factors and inflammation-inducing events (brain trauma, ischemia and infection) that influence the multifactorial etiology of LOAD. Both delirium and dementia encode a neuroinflammatory response that may describe the vulnerability of AD patients to additional cognitive deterioration. Subsequently, this response provokes a delirium occurrence related to a systemic inflammatory reaction. Abbreviations: Aβ, amyloid-β peptide; AD, Alzheimer’s disease; APOE4, apolipoprotein E4; APP, amyloid precursor protein; CLU, clusterin; CR1, complement receptor-1; PS1, presenilin-1; PS2, presenilin-2. From25.
The genetic findings on the etiology and pathogenesis of AD will continue to improve our knowledge on the pathological mechanisms underlying AD. Additionally, the search for further rare sequence variants in genes that prompt AD development will also aid in this understanding. Ultimately, these discoveries will support the development of new therapeutic approaches for preventing, stopping and even reversing AD course27.
Inflammation and immune response in Alzheimer’s disease
Many studies confirmed the presence of immune-related antigens and cells around amyloid plaques in the brains of individuals with AD, in addition to Aβ and tau protein aggregates28. Several
evidences support the involvement of inflammation and immune responses in the pathogenesis of psychiatric disorders such as dementia, depression, and schizophrenia29. The pathogenesis of AD is
also affected by immunological and inflammation mediated mechanisms. In the 1990s, studies addressing the activated complement factors, cytokines and a wide range of receptors in the brain of AD patients confirmed the association between immunological and inflammatory courses with this diseaseand that immune system-mediated activities promote AD pathogenesis30,31.
Aβ and Neuroinflammation
As previously described, the two main pathological hallmarks for AD, namely Aβ plaques and NFTs, are key players in AD course. According to the amyloid cascade hypothesis, Aβ accumulation in the brain — resulting from aberrant processing of APP or dysfunctional clearance of Aβ peptide — is the initiating event in AD. Additionally, every EOAD related mutation influences the formation or aggregation predisposition of the Aβ peptide. In fact, the majority of cases regarding EOAD are triggered by dominant genetic mutations, mainly inked to the PSEN1 and PSEN2 proteins, that constitute part of the γ-secretase complex involved in the proteolytic process of APP8.
Furthermore, the cleavage of APP occurs by two main ways: the non-amyloidogenic or the amyloidogenic pathway, although just the later leads to Aβ formation.
In the non-amyloidogenic via, that precludes Aβ formation, APP is initially cleaved by α-secretase, which generates a soluble N-terminal ectodomain named secreted amyloid precursor protein-α (sAPPα) - that is release from the cell surface -, and a truncated APP C-terminal fragment (CTF) (αCTF or C83). This fragment is subsequently cleaved by γ-secretase, leading to the secretion of a small non-pathogenic peptide p3 and a free APP intracellular domain (AICD), which is released into the cytosol32. On the other hand, in the amyloidogenic pathway, the β-secretase initiates Aβ
generation by shedding a large part of the ectodomain of APP (sAPPβ) and generating an APP CTF (βCTF or C99). This fragment is then cleaved by γ-secretase, generating Aβ and AICD. This final cleavage regulates the formation of the predominant Aβ40 or the more aggregation-prone and neurotoxic Aβ42 species (Figure 6)32,33
.
The amyloidogenic pathway products are the main pathological findings in AD35. Numerous
(soluble and insoluble) species and aggregation states of Aβ peptide can occur, comprising monomers, oligomers, protofibrils, fibrils and Aβ plaques. Further, the latest understandings about their biology indicate that these forms most likely have variable levels of pathogenic effects. Genetic findings sustain the hypothesis that atypical formation or increase of Aβ is a pathogenic feature in both EOAD and LOAD8 and reports confirmed the development of Aβ pathology on
transgenic mice with human APP mutations, as well as in cell lines36,37.
Additionally, the presence of Aβ can instigate microglial cells, which are the local macrophages of the CNS. These are the main cellular foundations of the innate immune system, not only by promoting the homeostasis regulation of other cells such as astrocytes and neurons in the CNS38,39, but also by remodeling the brain being involved in the removal of apoptotic neurons
and the reduction of tissue impairment40,41.
Figure 6: APP proteolytic processing. APP processing involves proteolytic cleavage by several secretases. The amyloidogenic pathway releases Aβ peptides through cleavage by β- and γ-secretases. The non-amyloidogenic pathway is initiated by α-secretase cleavage, which cuts the middle of the Aβ domain, resulting in the release of several soluble APP fragments. Abbreviations: Aβ, β-amyloid; APP-CTF, APP C-terminal fragment; AICD, APP intracellular domain. Reproduced from34.
Microglia ensure the adequate examination of their local microenvironment and protection by attacking pathogens via phagocytosis, generation of cytokines, oxyradicals, and initiation of the complement cascade42. Additionally, these CNS cells may also induce neural development,
influence vascular growth and dissipate toxic senile plaques by phagocytosis43. Microglia can
destroy soluble Aβ peptides via proteolytic degradation in endolysosomal compartments44 and
sphingolipid-modulated exosomes can activate microglia-mediated clearance of Aβ45. Conversely,
senescence of microglia and failure to correctly transform the microglial phenotype could impact the success of microglial amyloid clearance46,47,48.
In AD, Aβ endures chronic activation of stimulated microglia (due to the peptide’s accumulation), which produces a continuous level of inflammatory cytokines, chemokines, reactive oxygen and nitrogen species. In addition, these substances are responsible for preserving the activation of the stimulated cells, further discussed. Consequently, this mechanism leads to a vicious loop that eventually damages microglia. This damage also results from affected adjacent resident cells of the CNS, such as astrocytes, oligodendrocytes and neurons, and perhaps aggravated tau pathology as well, which lastly causes neurodegeneration and neuron loss. Moreover, if these courses extend over a continued period, it compels the microglia cells into a senescent, ‘burn-out’-like (dystrophic) phenotype, which is assumed to be permanent (Figure 7)49.
The amyloid cascade hypothesis supports a crucial pathogenic role for Aβ, a fundamental neurotoxic peptide in AD50. Immunological machinery and neuroinflammation triggered by A have
Clustered microglia and fibrous astrocytes surrounding extracellular deposits of Aβ plaques were mentioned in various publications. The accumulation of Aβ peptides, particularly aggregation-prone Aβ42 species, foment the formation of amyloid fibrils and Aβ oligomers in AD. Misfolded Aβ peptides can generate chemokines and similarly have a direct role as microglial attractants. For instance, chemokine receptors such as C-C chemokine receptor type-2 (CCR2), C-C chemokine receptor type-3 (CCR3), C-C chemokine receptor type-5 (CCR5), and CX3C chemokine receptor-1 (CX3CR1) exist in the microglia and brain of AD patients, and CXC chemokine receptor-2 (CXCR2) and CXC chemokine receptor-3 (CXCR33) are within the proximity of neuritic plaques53.
Microglia are not only observed within amyloid plaques, but they also connect to Aβ oligomers and Aβ fibrils by cell surface receptors, such as cluster of differentiation (CD) CD3654,
integrin-associated protein (IAP)/CD-47, α6β1 integrin55, scavenger receptor A-1 (SCARA1)56,
Toll-like receptors (TLRs)57 and associated receptors (e.g., CD14)58. In fact, the association between
receptor and Aβ fibrils promotes the in vitro phagocytosis by microglia cells of Aβ fibrils and the destruction of soluble Aβ species via extracellular proteases, neprilysin, and insulin-degrading
Figure 7: Microglial phenotypes prompted by Aβ. Generation of different microglia phenotypes stimulated by Aβ and other pathological protein deposits (alterations in the CNS, systemic or local inflammation, and mutations in genes encoding innate immune molecules). This process leads towards the formation of inflammatory cytokines and chemokines, that ultimately causes neurodegeneration, neuron loss and probably intensified tau pathology. From49.
enzyme (IDE)59.Successively, the connection of CD36 and TLRs 4 and 6 with Aβ peptides contribute
to the generation of proinflammatory cytokines and chemokines by microglia60.
Additionally, the complement system, a crucial feature of the innate immune system, comprises anaphylatoxins and opsonins, essential for Aβ clearance. In fact, genetic removal of complement factor C3 in mice has shown to increase Aβ deposition and neurodegeneration61. The
assembly of proteins from the complement system is also one role of microglia cells and astrocytes and reduced clearance of Aβ peptides by microglia adds to the pathogenesis of AD.
Chronic microglial activation is present in AD pathogenesis. For example, a positive feedback loop among inflammation and increased Aβ production leads to chronic, non-resolving inflammation62,63. Chronic inflammation leads to massive and continued release of
proinflammatory cytokines, chemokines, and other inflammation mediators. This process contributes to neuroinflammation and consequently to neuronal impairment. High levels of proinflammatory cytokines such as TNFα versus low levels of anti-inflammatory cytokine TGF-β induced the progression of dementia from MCI to AD64. Furthermore, microglial stimulation and
inflammatory reactions lead to activation of mitogen-activated protein kinase (MAPK) and intensifies hyperphosphorylation of tau protein in mouse models63,65. Cytokine-release impacts on
tau pathology, affects neuronal transport and injures neurons66.
On the other hand, microglia activation elevates nitric oxide (NO) and reactive oxygen species (ROS) production. In turn, ROS and reactive nitrogen species (RNS) can disrupt proteins, lipids, carbohydrates, deoxyribonucleic acid (DNA), and ribonucleic acid (RNA), which causes several forms of injury including cell death67. Oxidative stress can also intensify Aβ production and
aggregation process68. Additionally, inducible nitric oxide synthase (iNOS) is expressed in the brain
of patients with AD, and its genetic removal in mice is protective regarding AD course69.
Like microglia, astrocytes have a rapid reaction towards pathology, altering their morphology, antigenicity, and activity70. Astrocytes have an indispensable neuro-supportive role in
the brain and their functions include the release and reutilization of transmitters, ion homeostasis, energy metabolism control, synaptic remodeling, and modulation of oxidative stress.In fact, the disturbance of the neuro-supportive astrocyte features has highly damaging magnitudes for the CNS71.
Responsive astrocytes reside in peri-plaque areas of the brain in AD patients and in AD transgenic mouse models, with a characteristic form that resonant glial scarring - a mechanism by which the cells deliver a barrier among healthy tissue and areas of injury or infection72. The MCP1
Aβ plaques and acts as chemotactic for adult astrocytes. These astrocytes express receptors that associates with Aβ, including low density lipoprotein receptor-like protein, membrane-associated proteoglycans, and scavenger receptor-like receptors. Such system is responsible for the high accumulation of reactive astrocytes situated within Aβ deposits72.
Additionally, numerous studies concluded that plaque-localized, responsive astrocytes can degrade Aβ peptides73. Moreover, in Tg2576 transgenic mice this feature is related to insulin
degrading enzyme (IDE), crucial for Aβ degradation. IDE levels are higher in immunoreactive astrocytes around Aβ deposits74.
Astrocyte secretion of matrix metalloproteinases may also play a key role in extracellular clearance of Aβ75. Aβ disturbs astrocyte calcium homeostasis76 and elevates GFAP, a process
associated with degeneration of neurons in astrocyte-neuron co-cultures77. Responsive astrocytes
induce neuropathology via expression or overexpression of several of inflammation-related factors. For instances, S100 calcium-binding protein-b (S100b), a neurotropin that promotes neurite growth, is overexpressed in AD brain, due to countless dystrophic neurites within Aβ deposits78.
Further, Aβ can also lead to IL-1β, TNF-α, iNOS, and NO formation in astrocytes79.
Most of these results may be suggestive of modifications regarding astrocyte transcription factors. The proinflammatory cytokines IL-1β, IL-6, and TNF-α are regulated by nuclear kappa-B (NF-kB) and CCAAT-enhancer-binding proteins (C/EBP) transcription factors. Astrocyte NF-kB is also induced upon Aβ exposure, which augments IL-1β and IL-6 levels. Hence, the release of chemokines and adhesion molecules in astrocytes can be regulated by NF-Kb mechanisms, which possibly allows the incursion of peripheral leukocytes and adds to an inflammatory reaction in the brain80.
Immunity effects
The microglial activation process is very diverse and certainly influences the evolution of neurodegenerative diseases81.
Infiltrated blood macrophages from periphery and perivascular macrophages have an important role in the pathogenesis of AD82. For instance, activated perivascular macrophages
decrease the cerebral amyloid angiopathy and Aβ deposition in cerebral blood vessels and leptomeninges in a mouse model of AD83, an event involving the CCR2 chemokine, that is crucial for
Aβ clearance by perivascular macrophages84. Indeed, elevated infiltration levels from peripheral
macrophages were reported in the CNS of AD model mice85, which may partly contribute to Aβ
Several in vitro studies of oligodendrocyte toxicity has been described for Aβ25-3586,
Aβ4086,87, and Aβ4288. Oligodendrocytes express mRNAs and show immunoreactivity towards
complement components C1q, C1s, C2, C3, C4, C5, C6, C7, C8, and C9. This fact proposes that these may possibly be a major cause for complement augmentation in pathologically-vulnerable areas of the AD brain89. Complement stimulation takes place on oligodendrocytes via C1q association to
myelin oligodendrocyte protein and results from reduced levels of C1 inhibitor and membrane cofactor protein expressed by these cells.
Aβ can also impact at the white blood cells. Aβ fibrils function as an altered self-antigen, improving Aβ specific T lymphocytes. Blood vessels contiguous to Aβ fibrils display elevated levels of intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), which induces extravasation of stimulated Aβ specific T lymphocytes91.
Additionally, the population of lymphocytes was shown to be altered in peripheral immune system with low levels of peripheral T and B lymphocytes in AD patients92. In particular advanced
stages, it is possible to detect reduced levels of naive CD4+ T lymphocyte subpopulations, CD19+ B lymphocytes and naive B cells (immunoglobulin (Ig)-D+CD27−) and also increased double negative (IgD−CD27−) memory B cells detection. Further, a reduction of CD8+CD28− suppressor T lymphocytes and IL-10 production in AD was also reported, which suggests lower immunosuppressive skills93.
In AD, chronic inflammation promotes the reduction of naïve lymphocytes and the expression of altered chemokine receptors in lymphocytes. Furthermore, the elevated expression of pro-inflammatory chemokine receptors, C-C chemokine receptor type 6 (CCR6) and C-C chemokine receptor type 7 (CCR7) on B lymphocytes can contribute to chronic inflammatory condition94.
Clusterin
Clusterin (CLU) is an approximately 80kDa chaperone glycoprotein that can be dissociated in two 40kDa chains under reducing conditions. It is present in most of biological fluids95.
The human CLU gene is located on chromosome 8 and contains 9 exons presented with numerous transcriptional isoforms95,96. The mature protein matches a secretory form of clusterin
(sCLU), being a heterodimer with α and β subunits connected with five disulphide links. Individually, these subunit expresses three N-glycosylation locations with a carbohydrate heterogeneity, which improves the variety of glycoforms. In addition to sCLU that was initially identified, a nuclear form (nCLU) was described in human and murine cell lines exposed to many stress promoters such as radiations, heat shock, drugs, as well as in carcinoma cells. Furthermore, nCLU assembly was also observed in other conditions such as ethanol-induced cell death and hypoxia in the neonatal rodent brain97.
Contrarily to sCLU cell survival effects, nCLU is related to cell death by apoptosis. The sCLU is extensively expressed in the body, being mostly secreted in CSF98. Elevated plasma clusterin in
healthy centenarians (older than 105 years) suggests a potential function for this protein through longevity99. These mechanisms are still not completely understood, although additional reports
acknowledged the involvement of other molecules that also function as binding ligands of sCLU such as complement factors, immunoglobulins, leptin, Aβ, alpha-synuclein (α-synuclein), prion protein, transforming growth factor beta (TGFβ)-receptors and stressed unfolded proteins100,101.
Clusterin in Alzheimer’s disease
Clusterin is associated with lipid transport and metabolism, membrane protection, cell-cell interaction, apoptosis, and complement regulation. Regarding the complement system, clusterin controls the membrane attack complex and limits complement stimulation related to inflammatory reaction102.
Under normal conditions, some neuronal subpopulations, particularly at the pontobulbar and spinal cord motor nuclei have high expression levels of CLU. Moreover, latest investigations in postmortem human brain reveled plenty populations of neurons expressing low levels of clusterin mRNA in the neocortex. CLU is also present within a few subpopulations of astrocytes, specifically over the hippocampal and in some neocortical subdivisions103.
Nowadays, it is known that CLU levels in the brain fluctuate throughout healthy aging. In line with this evidence significant age-related difference in clusterin expression expression were observed in glial cells 105.
Reduced levels of CLU in high-density lipoproteins is linked to insulin resistance and obesity, which are also known risk factors for AD. In AD, several studies reported the co-localization of CLU in NFTs and extracellular Aβ deposits. Further, GWAS and meta-analyses indicated CLU as one of the LOAD susceptibility genes.These studies also described that increased clusterin plasma levels may be associated with brain atrophy, disease severity, and disease progression106.
In AD, the CLU variants may obstruct the usual function of clusterin as a protective factor against oxidative stress, by inhibiting the normal role of clusterin as a sensor and chaperone of ROS. Among different reported SNPsthe variant rs11136000 has a particular connection with plasma CLU levels and contributes to AD with similar genetic risk effects in both the Asian and Caucasian populations. The T-allele is assumed to be associated with reduced risk in AD development and better cognitive performance, whereas the C-allele represents a risk factor for AD development. It was reported that homozygous expression of the C-allele facilitates neural hyperactivity in frontal and posterior cingulate cortices as well as hippocampus, throughout a visual working memory task, a hyperactivation that at long term can be detrimental107,108,109.
Furthermore, additional findings revealed reduced or even absent hippocampal–prefrontal connectivity during memory retrieval in AD patients and also in subjects suffering from MCI110. For
the risk variant, a C-allele dosage-dependent variation of the efficient combination in hippocampus and prefrontal cortex during episodic memory tasks was observed applying genetic and imaging methods on healthy subjects (Figure 8)103.
Further reports showed altered white matter integrity in healthy carriers of the rs11136000 risk variant, which translates to neurodevelopmental susceptibility111. Additionally, young healthy
carriers of the C-allele presented a diminished fractional anisotropy, caused by myelin integrity loss rather than axonal injuries, using diffusion tensor MRI imaging. Microstructural alterations in long interhemispheric and intrahemispheric associations such as the corpus callosum, the fornix and the cingulum, are related to a prodromal vulnerability in AD112,113.
As mentioned, individuals expressing the genotype related to greater risk (CC homozygous genotype)have higher neural activation in a task related network. A possible explanation would be that carriers of the protective T-allele would achieve an equivalent level of task performance, although with less brain stimulation and that greater cognitive effort is necessary in at-risk
Figure 8: Altered functional coupling in carriers of rs11136000 risk variant for CLU. A: carriers of the C-risk allele reveal considerably reduced allele dosage-dependent coupling of the right dorsolateral prefrontal cortex (DLPFC) with the right hippocampus seed region through recall. Each red dot indicates size of effect in one subject and reproduces the connectivity among right DLPFC and right hippocampal seed area. B: carriers of the C-risk allele reveal considerably reduced allele dosage-dependent coupling of the right dorsolateral DLPFC with the left hippocampus seed region through recognition. Each red dot indicates size of effect in one subject and reproduces connectivity among the right DLPFC and the right hippocampal seed area Adapted from103.
individuals to complete equivalent levels of performance102. Moreover, additional reports
described that carriers of this recently discovered rs11136000 AD risk variant of CLU, had longitudinal increases in neural activity resting state in brain regions critical to memory processes, in order to maintain their physiological function in cognitively normal at-risk individuals. Supporting these supposed compensatory boosts in neural response in some individuals may be the threshold beyond which primary cognitive damage initiates, setting their conversion from MCI to AD114.
Increased hippocampal stimulation throughout memory processes in MCI individuals has also been observed in rs11136000 risk carriers, which may ultimately express earlier rates of decline in memory performance comparative to non-carriers at the pre-symptomatic stages of AD115,116.
Besides, increased levels of CLU were identified on cerebrospinal fluid of AD patients, which may represent a relevant peripheral marker associated with disease development.
In vitro and in vivo experimental models indicated a potential linkage between CLU and Aβ
aggregation and clearance. Clusterin levels improve the degree of Aβ clearance and efflux via the blood brain barrier117. Furthermore, extracellular sCLU mutually limits Aβ40 monomers aggregation
and also the toxic and inflammatory effects ofoligomers by sequestration previous to peptides degradation. In addition, a recent meta-analysis demonstrates that sporadic coding variants affecting sCLU β-chain are mostly observed in AD118.
Complement receptor-1
Complement receptor-1 (CR1), also known as C3b/C4b receptor or CD35, is situated on
chromosome 1q32 and encodes CR1 protein. It is associated with the complement response and is part of the regulators of the complement activation (RCA) family of proteins119.
The largest production of complement proteins occurs in the liver, but is also highly present in the serum. Nevertheless, glial cells and neurons in the CNS can synthesize these proteins as well during injury and neurodegeneration120,121-125.
The complement system identifies molecular arrangements on pathogens or molecular patterns associated with damaged tissues and dying cells. Recognition may occur via C1q or mannose binding proteins, which have collagen-like receptor binding domains, or over interfaces with the multifunctional protein C3. After being triggered, a cascade of several proteins from the complement pathway can attract and stimulate immune cells, amplify antigen-specific immune responses, induce phagocytosis, support complement-mediated cytolysis by the membrane attack complex (MAC), and control cell proliferation and differentiation120.
Furthermore, both classical and alternative pathways activation can form catalytic multiproteins, namely the C3 convertases, producing two proteolytic C3 fragments: C3a, which is associated with inflammation and phagocyte recruitment, and C3b that could trigger the lytic pathway related to C5, C6, C7, C8, and C9, which ultimately generates MAC, C5b-9, and may induce cytotoxicity. This pathway also leads to C5a activation, an additional small proinflammatory peptide. Then, C3b binding to specific molecules, by a mechanism entitled opsonization, facilitating phagocytosis by a number of cells through their surface expression of complement receptors (Figure 9)111.
Complement receptor 1 in Alzheimer’s disease
Many polymorphisms for CR1 have been studied, demonstrating that SNPs may affect the expression of CR1 molecules on the cell surface119. In AD, these SNPs on CR1 are reported to be
associated with greater risk for disease development120. In fact, Aβ can activate the complement
system121 and large GWAS confirmed that rs3818361 within the CR1 locus is associated with LOAD
susceptibility in Caucasians122. The relationship between CR1 markers and LOAD in European and
American population is additionally confirmed by associations between disease and the rs3818361 variant26,27. This variant is also associated with neuroimaging measures and neuritic plaques burden
in AD brains.
Studies have explored inter-group variances in brain amyloid burden among risk (AG heterozygous genotype/AA homozygous genotype) and non-risk (GG homozygous genotype) carriers of the AD rs3818361 SNP in CR1 gene. Recent GWAS results showed a larger risk in carriers of the A-risk allele of this SNP in AD individuals. Significantly greater variance in brain amyloid deposition was observed in the non-risk group (GG homozygous genotype), which appeared to be associated in part with APOE genotype120. Consequently, GG homozygous genotype and APOE ε4
carriers had more amyloid deposition in numerous brain areas, comparative to APOEε4 non-carriers.
Figure 9: Schematic interaction of the complement system components. Abbreviations: complement proteins, C1q, C3, C3b, C3a, C5, C5a, C5b, C6, C7, C8, C9; membrane attack complex, MAC. From111.
Although the CR1 A-risk allele was linked to reduced fibrillar amyloid in non-demented individuals, these authors have also observed elevated brain amyloid levels in carriers of the APOEε4 allele relative to non-carriers in cognitively normal older individuals120. Additional studies
confirmed that individuals expressing GG homozygous genotype and APOEε4 positive results exhibited reduced episodic memory, which is a common endophenotype of LOAD. This indicates that there is a greater susceptibility for AD development in individuals caring both the risk variant rs3818361 and the APOEε4 allele26.
Additionally, CR1 is pointed as a vital AD susceptibility gene, since the expression of complement factors are upregulated in affected regions of AD brains126.
CR1 protein regulates the complement activity and high complement cascade activity can exacerbate AD pathology. Essentially, the presence of CR1 on phagocytic cells induces them to phagocyte specific particles, by stimulating the complement reaction. Of note, high CR1 expression levels are related to elevated cognitive deterioration in AD.Further, Aβ clearance in the brain is associated with CR1 expression127. Clearance of plasma Aβ42 is dependent on binding of CR1, which
therefore leads to glial activation, ultimately promoting neural degradation and chronic inflammatory cascade activation127,128. CR1 protein expression may modulate microglia, leading to
an amplified clearance rate of immune complexes129. Since genetic variance in CR1 modifies the
volume of Aβ clearance by the immune response, inadequate clearance of Aβ and amplified chronic inflammation by CR1 polymorphisms can contribute to AD pathogenesis130.
Aims of the Thesis
Both CLU and CR1 are two of the most common genes from Alzgene database. These genes comprise risk alleles polymorphisms involved in AD development and pathogenesis, but limited evidences are known about the molecular mechanisms involved.
CLU encodes for the protein clusterin, which is highly expressed in the brains of individuals
with AD. It can be related to Aβ aggregation, and complement response activation. In fact, in individuals with MCI and AD, elevated plasma clusterin is correlated with a more rapid cognitive decline. On the other hand, CR1 is a complement receptor present in the cerebral cortex. Emerging evidence supports the notion that the complement cascade is involved in neurodegeneration process prompted by astrocyte-mediated opsonisation by CR1. Thus, CLU and CR1 have emerged as very interesting targets for AD.
The aim of this thesis was to address both inflammation-related gene variants in a dementia subpopulation group from a primary care-based cohort previously established by the Neuroscience and Signalling group. Hence the specific aims of this thesis were to:
- Perform a literature survey and a bioinformatic analysis on AD inflammation related-genes;
- Amplify the polymorphic regions rs11136000 and rs3818361 of CLU and CR1 genes respectively, from blood samples of the study group;
- Determine the genotype and the allelic frequency of the polymorphic regions rs11136000 and rs3818361 of CLU and CR1 in the study group;
Materials and Methods
Bioinformatics analyses
Literature survey on AD risk-related genes and construction of a
protein-protein interaction network
The genes identified for this study were collected from a literature survey of review articles published between 1/1/2015 to 31/12/2015 at PubMed database131 (using the key words
Alzheimer’s disease and risk genes, article types), that complemented the data available on AD risk related genes. Only the interactions that were curated in Homo sapiens have been included.
The genes were identified by their own ID: gene symbol (www.genecards.org/)132 and
UniProtKB identifier (www.uniprot.org/)133. The gene symbol identifier of the collected proteins
was loaded into STRING database (http://string-db.org)134 to generate a protein-protein interaction
network. The network was generated using data from “Gene Ontology (GO) Biological Processes”, “GO Molecular Functions” and “GO Cellular Components” as a network enrichment method. Markov Cluster algorithm (MCL) was also applied in order to calculate the powers of the associated adjacency of this network.
Gene ontology analysis
The analysis (adjusted P-value < 0.05) was carried out using STRING database according to GO annotation. Data regarding CLU and CR1 genes were collected for ‘biological processes’, ‘molecular functions’ and ‘cellular component’, particularly focusing on GO in which these genes were involved.
Characterization of the study group
Our pilot study group included a population of 63 individuals from a cross-sectional population-based study on a Portuguese population from the Aveiro region: 32 of which were individuals with dementia, based on cognitive evaluation tests, herein referred as “Putative AD” group and 31 individuals that forms the Controls group. This is part of a project approved by the ethics committee for health of the central regional administration in Coimbra (Comissão de Ética para a Saúde da ARS Centro, protocol 0128804-04.04.2012), and by the National Committee for Data Protection.
These 63 individuals represent a subgroup of a total of 590 individuals (568 fulfilled the inclusion criteria), nominated by primary care-based cohort (pcb-Cohort). Individuals undertook a
survey and were submitted to several cognitive evaluations and dementia screening tests during an interview, regardless of the clinical diagnosis. Screening tests comprised Clinical Dementia Rate (CDR), Mini Mental State Examination (MMSE), Geriatric Depression Scale (GDS), Katz of Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL). A subset of individuals that were both CDR positive (CDR≥1, that represents a dementia cognitive rate from moderate to severe) and MMSE positive, were designated as “Putative AD” group, which in fact included individuals clinically diagnosed for AD. The “Putative AD” group came from a dementia risk group of 68 individuals based on CDR classification (Martins et al. 2016, Dementia geriatric cognitive disorder, in press), that were also MMSE positive. Controls and “Putative AD” individuals were age and sex matched. All participants provided written informed consent for the study. The age and gender variables are shown on Table 2.
Table 2. Representation of the gender and age variables between the “Putative AD” group and the Controls group.
GENDER AGE MEAN AGE
N % INTERVAL N % CONTROLS MALE 76 10 32% 50-64 3 30 65-74 1 10 ≥75 6 60 FEMALE 21 68% 50-64 2 10 65-74 4 19 ≥75 1 5 71
“PUTATIVE AD” MALE 77
10 31% 50-64 2 20 65-74 2 20 ≥75 6 60 FEMALE 22 69% 50-64 2 9 65-74 4 18 ≥75 1 6 73
According to the gender variable, the entire study group was mainly represented by female individuals (69% for “Putative AD” group and 68% for the Controls group, against 31% and 32% for males from the “Putative AD” group and Controls group, respectively). A large portion of the samples were from female individuals over 75 years old for both “Putative AD” and for Controls groups, comparative to male individuals. There were no significant differences regarding the mean ages for both analysed groups.
PCR analyses
Blood samples were collected in EDTA-tubes according to standard procedures and were promptly aliquoted and frozen at -80°C, upon arrival in the laboratory.
PCR technique is a scientific tool used to replicate DNA from a template, generating numerous copies of a specific sequence. This technique was used to amplify a specific portion of
CLU and CR1 polymorphic regions, respectively rs11136000 (with 685bp) and rs3818361 (with
664bp) from each patient. For the PCR reaction, Phusion Blood Direct PCR Master Mix (ThermoFisher Scientific, USA) was employed according to the manufacturer’s instructions. 1µl of whole blood was used in the amplification of each gene and no previous gDNA extraction was required.
Primers were designed as previously described169 and tested for uniqueness using the NCBI BLAST W search engine (Table 3).
Table 3: Sequence of the primers and the concentration used for the genotyping of the CLU and CR1 polymorphism135. Gene SNP ID Primer name Sequence (5’-3’) Concentration of primer in AS-PCR (μM) CLU rs11136000 CLU-Fw CLU-Rv CCTGGCTTAAAGAATCCACTCATC CAGGGGATTCCTTTGAGATAGAGT 0.1μM 0.1μM CR1 rs3818361 CR1-Fw CR1-Rv TTCAACTACTGGTTATGGAGCA CACTCACCCTTCATCGCAAA 0.1μM 0.1μM
PCR technique was performed in a final volume of 20 µL, per gene reaction. The PCR round comprised 10 µL of 2XPhusion Blood II DNA Polymerase Master Mix (ThermoFisher Scientific, USA), 1μL of primer forward (Fw) (Eurogentec, Belgium) and 1µL of primer reverse (Rv) (Eurogentec, Belgium) of each corresponding gene, 1 μLof whole blood and 7 μL of ultrapure DNase/RNase-free distilled H2O (ThermoFisher Scientific, USA).
The PCR program included an initial denaturation step at 98°C for 5min, followed by 35 cycles of amplification at 94°C for 1min, 63°C for 30sec and 72°C for 46sec. The final extension was performed at 72°C for 5min.
Following amplification, the PCR products were centrifuged at 1000×g during 1 to 3 min to sediment all blood residues and DNA samples were precipitated as described below.
1/10 of sodium acetate (pH 3M, pH 5.2), followed by 2.5 volumes of 100% Ethanol were added to the solution, which was then mixed and placed at -20°C overnight. Afterwards, the tubes were centrifuged at 14000rpm for 20 min at 4⁰C. The supernatant was discarded and 100 μL of Ethanol 70% was added to the remaining pellet. The tubes were placed at -20°C for 20 min, and later on centrifuged at 14000rpm during 5 min. The supernatant was discarded and the DNA pellet was dried out completely at 37°C. Finally, 13 μL of ultrapure DNase/RNase-free distilled H2O was
added to each tube in order to dissolve the pellet. The purified DNA fragments were stored at -20°C for long term storage and/or 4°C for short term storage.
In order to verify the successful precipitation of the DNA fragments, 2 µL of each sample were analysed by agarose gel electrophoresis. DNA was precipitated from 10 µL of the previous amplified PCR product.
Agarose gel electrophoresis is a technique used for separation of DNA fragments. As DNA is a negatively charged molecule, it migrates from cathode to anode by the porous matrix of agarose, via electric field. The separation process occurs according to the fragment size, leading to a faster migration by the shorter molecules. The percentage of the agarose gel depends on the DNA band size that we intend to examine.
Subsequently to the precipitation of the DNA products, conventional electrophoresis was executed at 100V for 35 min in 1XTris-acetate-EDTA buffer on 1.5% agarose gel stained with 4 µL of GreenSafe Premium® (Nzytech, Portugal)/100ml of agarose (Nzytech, Portugal) solution. From the PCR products 2 µL was used to run on the gel. Loading buffer (LB) was added in a ratio of 1:6 to increase sample density. The 1Kb Plus DNA Ladder (ThermoFisher Scientific, USA) was used as a molecular weight marker. The purified DNA fragments were visualized under ultraviolet (UV) light and further analysed with ImageLab 5.2.1 (Bio-Rad Laboratories, Inc., USA) software.
After DNA precipitation, 10 ml of the product was analyzed by Sanger Sequencing, using the previous primers to allow CLU and CR1 polymorphisms detection.
Sequencing analysis
Sanger sequencing technique uses a DNA polymerase that copy single-stranded DNA templates by assembling nucleotides to a growing chain (extension product). This process occurs in the 5' -> 3' direction. At the 3' end of a primer starts the chain elongation, being the place were an oligonucleotide anneals to the template. The product extension occurs by additional deoxynucleotides incorporation by complementary to the template. The extension product grows by the formation of a phosphodiester bridge among the 3'-hydroxyl group on the primer and the 5'-phosphate group of the incoming deoxynucleotide, which enables the development of the extension product.
Further, DNA polymerases can also add analogues of nucleotide bases, as the dideoxy method of DNA sequencing that uses 2',3'-dideoxynucleotides as substrates. When dideoxynucleotides are combined at the 3' end of the developing chain, chain elongation is completed selectively at A, C, G, or T. After the binding of the dideoxynucleotide, the chain is absent of a 3'-hydroxyl group therefore further elongation of the chain is prevented.
